摘要
由于受到干扰或自身稳定性等因素的影响,在试飞中雷达或光电经纬仪会出现连片野值,影响测试数据的准确性。为此,采用基于Bayes定理的抗野值Kalman滤波算法,根据测量误差建立了受污分布模型,得到后验概率,据此确定预报方差的权值,以降低连片野值的影响。经仿真试验和飞行试验验证,该方法稳定有效,实时性强。已经应用在试飞测试工作中。
Due to influence of interference or their stability and other factors, radar or theodolite may appear contiguous outliers in test flight, the accuracy of test data may be affected. Based on Kalman filter algorithm outliers Bayes theorem, according to measurement error, contaminated distribution model is established, the posteriori probability is gotten, weights of forecast variance are determined to reduce the impact of contiguous outliers. The simulation tests and flight tests verify the method is stable and effective, real-time, which has been used in flight test work.
出处
《测控技术》
CSCD
北大核心
2014年第9期26-28,32,共4页
Measurement & Control Technology
基金
航空基金资助项目(2010ZD3004)
